Life Prediction of Rolling Bearing Based on Bidirectional GRU

Abstract

Rolling bearing is a kind of precision mechanical component bearing that changes the sliding friction between the running shaft and the shaft seat into rolling friction, so as to reduce the friction loss. It is a very important part of mechanical equipment, and its life prediction is of great significance. In this regard, a residual life prediction method based on bi-directional Gate recurrent unit is proposed. Firstly, the time domain, frequency domain and time-frequency domain features are screened, and three evaluation indexes are defined to quantitatively evaluate the effect of feature parameters characterizing the bearing degradation process. The sensitive feature set is screened. A bi-directional Gate recurrent unit network is built, and the sensitive feature set is used as input, The normalized single point life value is changed to segment life value as a label to train the neural network, and finally the life prediction of rolling bearing is realized. Finally, it is verified on the public data set that this method can accurately predict the remaining life of rolling bearings.